Project Summary/Abstract There were an estimated 10.4 million tuberculosis (TB) cases, and 170,000 TB pediatric deaths in 2015, with >95% of TB deaths occur in low- and middle-income countries (LMIC). Rapid diagnosis is essential for improved TB outcomes, but diagnosis is more difficult in children, since children are less likely to exhibit typical TB symptoms and have paucibacillary (few bacteria) TB cases that are more difficult to detect by current methods, while parents are often resistant to stressful TB sample collection procedures. We therefore propose to explore a novel rapid, accurate, low-cost and easy-to-use mobile phone-based blood test for pediatric TB diagnosis in LMIC settings. Our novel detection method rapidly quantifies circulating extracellular vesicles (EVs) derived from host cells infected with the TB pathogen Mycobacterium tuberculosis (Mtb). Our preliminary data indicate that we can rapidly quantitate EVs carrying Mtb biomarkers, such as lipoarabinomannan and LpqH, in pediatric TB patient blood samples using a mobile-phone-based dark-field microscope (MDFM) platform. These findings support our long-term goal to develop a low-cost system to improve pediatric TB diagnosis in LMIC settings. We also anticipate that this system will allow rapid treatment monitoring to improve therapy and reduce exposure of pediatric patients to toxic anti-TB drugs. To achieve these outcomes we propose to pursue three goals. We collected our preliminary data with a MDFM that imaged transmitted light, but plan to design, develop, and validate the Aim 1 compact MDFM with a reflected light path to decrease the size of the device, simplify imagery, and increase the physical stability of the system. We will also develop a user-friendly application for image capture and data analysis on this platform. These features should markedly increase its feasibility for use in LMIC settings. Aim 2 will optimize assay performance for LMIC settings by redesigning assay materials for long-term storage under ambient conditions, and by determining optimal incubation times for different LMIC ambient temperature ranges. Aim 3 will analyze candidate Mtb-EV markers, including lipoarabinomannan and LpqH, build and a diagnostic model based on a multiple logistic regression algorithm, and validate diagnostic thresholds in separate validation cohort of pediatric TB patients and healthy controls with traceable clinical information. It will also compare the diagnostic performance of this multi-marker Mtb-EV assay to traditional diagnostic methods to evaluate its relative diagnostic performance. Our approach offers several innovative features valuable for pediatric TB control. It quantifies stable TB biomarkers on EVs in serum (1L), rather than detecting Mtb in difficult to obtain, variable, non-quantitative and infectious biopsies. It employs a novel, compact MDFM and a nanoparticle-based end-point assay that is stable at ambient conditions. Finally, the ability of this approach to quantitate Mtb-EVs should allow rapid evaluation of disease severity for real-time monitoring of treatment efficacy and cures to minimize toxic drug exposure times.